The potential of genomics for infectious disease forecasting

Genomic surveillance as a scalable framework for precision phage therapy against antibiotic-resistant pathogens

Phage therapy is gaining increasing interest in the fight against critically antibiotic-resistant nosocomial pathogens. However, the narrow host range of bacteriophages hampers the development of broadly effective phage therapeutics and demands precision approaches. Here, we combine large-scale phylogeographic analysis with high-throughput phage typing to guide the development of precision phage cocktails targeting carbapenem-resistant Acinetobacter baumannii, a top-priority pathogen. Our analysis reveals that a few strain types dominate infections in each world region, with their geographical distribution remaining stable within 6 years. As we demonstrate in Eastern Europe, this spatiotemporal distribution enables preemptive preparation of region-specific phage collections that target most local infections. Finally, we showcase the efficacy of phage cocktails against prevalent strain types using in vitro and animal infection models. Ultimately, genomic surveillance identifies patients benefiting from the same phages across geographical scales, thus providing a scalable framework for precision phage therapy.

Implementing a national programme of pathogen genomics for public health: the Australian Pathogen Genomics Program (AusPathoGen)

Delivering large-scale routine pathogen genomics surveillance for public health is of considerable interest, although translational research models that promote national-level implementation are not well defined. We describe the development and deployment of the Australian Pathogen Genomics Program (AusPathoGen), a comprehensive national partnership between academia, public health laboratories, and public health agencies that commenced in January, 2021. Successfully establishing and delivering a national programme requires inclusive and transparent collaboration between stakeholders, defined and clear focus on public health priorities, and support for strengthening national genomics capacity. Major enablers for delivering such a programme include technical solutions for data integration and analysis, such as the genomics surveillance platform AusTrakka, standard bioinformatic analysis methods, and national ethics and data sharing agreements that promote nationally integrated surveillance systems. Training of public health officials to interpret and act on genomic data is crucial, and evaluation and cost-effectiveness programmes will provide a benchmark and evidence for sustainable investment in genomics nationally and globally.

Pathogen genomic surveillance status among lower resource settings in Asia

Asia remains vulnerable to new and emerging infectious diseases. Understanding how to improve next generation sequencing (NGS) use in pathogen surveillance is an urgent priority for regional health security. Here we developed a pathogen genomic surveillance assessment framework to assess capacity in low-resource settings in South and Southeast Asia. Data collected between June 2022 and March 2023 from 42 institutions in 13 countries showed pathogen genomics capacity exists, but use is limited and under-resourced. All countries had NGS capacity and seven countries had strategic plans integrating pathogen genomics into wider surveillance efforts. Several pathogens were prioritized for human surveillance, but NGS application to environmental and human-animal interface surveillance was limited. Barriers to NGS implementation include reliance on external funding, supply chain challenges, trained personnel shortages and limited quality assurance mechanisms. Coordinated efforts are required to support national planning, address capacity gaps, enhance quality assurance and facilitate data sharing for decision making.

The genetic architecture of protein stability

There are more ways to synthesize a 100-amino acid (aa) protein (20100) than there are atoms in the universe. Only a very small fraction of such a vast sequence space can ever be experimentally or computationally surveyed. Deep neural networks are increasingly being used to navigate high-dimensional sequence spaces1. However, these models are extremely complicated. Here, by experimentally sampling from sequence spaces larger than 1010, we show that the genetic architecture of at least some proteins is remarkably simple, allowing accurate genetic prediction in high-dimensional sequence spaces with fully interpretable energy models. These models capture the nonlinear relationships between free energies and phenotypes but otherwise consist of additive free energy changes with a small contribution from pairwise energetic couplings. These energetic couplings are sparse and associated with structural contacts and backbone proximity. Our results indicate that protein genetics is actually both rather simple and intelligible.

Digitized Kinetic Analysis Enhances Genotyping Capacity of CRISPR-Based Biosensing

CRISPR/Cas systems have been widely employed for nucleic acid biosensing and have been further advanced for mutation detection by virtue of the sequence specificity of crRNA. However, existing CRISPR-based genotyping methods are limited by the mismatch tolerance of Cas effectors, necessitating a comprehensive screening of crRNAs to effectively distinguish between wild-type and point-mutated sequences. To circumvent the limitation of conventional CRISPR-based genotyping, here, we introduce Single-Molecule kinetic Analysis via a Real-Time digital CRISPR/Cas12a-assisted assay (SMART-dCRISPR). SMART-dCRISPR leverages the differential kinetics of the signal increase in CRISPR/Cas systems, which is modulated by the complementarity between crRNA and the target sequence. It employs single-molecule digital measurements to discern mutations based on kinetic profiles that could otherwise be obscured by variations in the target concentrations. We applied SMART-dCRISPR to genotype notable mutations in SARS-CoV-2, point mutation (K417N) and deletion (69/70DEL), successfully distinguishing wild-type, Omicron BA.1, and Omicron BA.2 SARS-CoV-2 strains from clinical nasopharyngeal/nasal swab samples. Additionally, we introduced a portable digital real-time sensing device to streamline SMART-dCRISPR and enhance its practicality for point-of-care settings. The combination of a rapid and sensitive isothermal CRISPR-based assay with single-molecule kinetic analysis in a portable format significantly enhances the versatility of CRISPR-based nucleic acid biosensing and genotyping.

RSV-GenoScan: An automated pipeline for whole-genome human respiratory syncytial virus (RSV) sequence analysis

BACKGROUND

Advances in high-throughput sequencing (HTS) technologies and reductions in sequencing costs have revolutionised the study of genomics and molecular biology by making whole-genome sequencing (WGS) accessible to many laboratories. However, the analysis of WGS data requires significant computational effort, which is the major drawback in implementing WGS as a routine laboratory technique.

OBJECTIVE

Automated pipelines have been developed to overcome this issue, but they do not exist for all organisms. This is the case for human respiratory syncytial virus (RSV), which is a leading cause of lower respiratory tract infections in infants, the elderly, and immunocompromised adults.

RESULTS

We present RSV-GenoScan, a fast and easy-to-use pipeline for WGS analysis of RSV generated by HTS on Illumina or Nanopore platforms. RSV-GenoScan automates the WGS analysis steps directly from the raw sequence data. The pipeline filters the sequence data, maps the reads to the RSV reference genomes, generates a consensus sequence, identifies the RSV subgroup, and lists amino acid mutations, insertions and deletions in the F and G viral genes. This enables the rapid identification of mutations in these coding genes that are known to confer resistance to monoclonal antibodies.

AVAILABILITY

RSV-GenoScan is freely available at https://github.com/AlexandreD-bio/RSV-GenoScan.

Evaluating the performance of Plasmodium falciparum genetic metrics for inferring National Malaria Control Programme reported incidence in Senegal

BACKGROUND

Genetic surveillance of the Plasmodium falciparum parasite shows great promise for helping National Malaria Control Programmes (NMCPs) assess parasite transmission. Genetic metrics such as the frequency of polygenomic (multiple strain) infections, genetic clones, and the complexity of infection (COI, number of strains per infection) are correlated with transmission intensity. However, despite these correlations, it is unclear whether genetic metrics alone are sufficient to estimate clinical incidence.

METHODS

This study examined parasites from 3147 clinical infections sampled between the years 2012-2020 through passive case detection (PCD) across 16 clinic sites spread throughout Senegal. Samples were genotyped with a 24 single nucleotide polymorphism (SNP) molecular barcode that detects parasite strains, distinguishes polygenomic (multiple strain) from monogenomic (single strain) infections, and identifies clonal infections. To determine whether genetic signals can predict incidence, a series of Poisson generalized linear mixed-effects models were constructed to predict the incidence level at each clinical site from a set of genetic metrics designed to measure parasite clonality, superinfection, and co-transmission rates.

RESULTS

Model-predicted incidence was compared with the reported standard incidence data determined by the NMCP for each clinic and found that parasite genetic metrics generally correlated with reported incidence, with departures from expected values at very low annual incidence (< 10/1000/annual [‰]).

CONCLUSIONS

When transmission is greater than 10 cases per 1000 annual parasite incidence (annual incidence > 10‰), parasite genetics can be used to accurately infer incidence and is consistent with superinfection-based hypotheses of malaria transmission. When transmission was < 10‰, many of the correlations between parasite genetics and incidence were reversed, which may reflect the disproportionate impact of importation and focal transmission on parasite genetics when local transmission levels are low.

Phylogenomic early warning signals for SARS-CoV-2 epidemic waves

BACKGROUND

Epidemic waves of coronavirus disease 2019 (COVID-19) infections have often been associated with the emergence of novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants. Rapid detection of growing genomic variants can therefore serve as a predictor of future waves, enabling timely implementation of countermeasures such as non-pharmaceutical interventions (social distancing), additional vaccination (booster campaigns), or healthcare capacity adjustments. The large amount of SARS-CoV-2 genomic sequence data produced during the pandemic has provided a unique opportunity to explore the utility of these data for generating early warning signals (EWS).

METHODS

We developed an analytical pipeline (Transmission Fitness Polymorphism Scanner - designated in an R package mrc-ide/tfpscanner) for systematically exploring all clades within a SARS-CoV-2 virus phylogeny to detect variants showing unusually high growth rates. We investigated the use of these cluster growth rates as the basis for a variety of statistical time series to use as leading indicators for the epidemic waves in the UK during the pandemic between August 2020 and March 2022. We also compared the performance of these phylogeny-derived leading indicators with a range of non-phylogeny-derived leading indicators. Our experiments simulated data generation and real-time analysis.

FINDINGS

Using phylogenomic analysis, we identified leading indicators that would have generated EWS ahead of significant increases in COVID-19 hospitalisations in the UK between August 2020 and March 2022. Our results also show that EWS lead time is sensitive to the threshold set for the number of false positive (FP) EWS. It is often possible to generate longer EWS lead times if more FP EWS are tolerated. On the basis of maximising lead time and minimising the number of FP EWS, the best performing leading indicators that we identified, amongst a set of 1.4 million, were the maximum logistic growth rate (LGR) amongst clusters of the dominant Pango lineage and the mean simple LGR across a broader set of clusters. In the case of the former, the time between the EWS and wave inflection points (a conservative measure of wave start dates) for the seven waves ranged between a 20-day lead time and a 7-day lag, with a mean lead time of 5.4 days. The maximum number of FP EWS generated prior to a true positive (TP) EWS was two and this only occurred for two of the seven waves in the period. The mean simple LGR amongst a broader set of clusters also performed well in terms of lead time but with slightly more FP EWS.

INTERPRETATION

As a result of the significant surveillance effort during the pandemic, early detection of SARS-CoV-2 variants of concern Alpha, Delta, and Omicron provided some of the first examples where timely detection and characterisation of pathogen variants has been used to tailor public health response. The success of our method in generating early warning signals based on phylogenomic analysis for SARS-CoV-2 in the UK may make it a worthwhile addition to existing surveillance strategies. In addition, the method may be translatable to other countries and/or regions, and to other pathogens with large-scale and rapid genomic surveillance.

FUNDING

This research was funded in whole, or in part, by the Wellcome Trust (220885_Z_20_Z). For the purpose of open access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. KOD, OB, VBF and EMV acknowledge funding from the MRC Centre for Global Infectious Disease Analysis (reference MR/X020258/1), jointly funded by the UK Medical Research Council (MRC) and the UK Foreign, Commonwealth & Development Office (FCDO), under the MRC/FCDO Concordat agreement and is also part of the EDCTP2 programme supported by the European Union. RMC acknowledges funding from the Wellcome Trust Collaborators Award (206298/Z/17/Z).

Phenotypic drug-susceptibility profiles and genetic analysis based on whole-genome sequencing of Mycobacterium avium complex isolates in Thailand

Mycobacterium avium complex (MAC) infections are a significant clinical challenge. Determining drug-susceptibility profiles and the genetic basis of drug resistance is crucial for guiding effective treatment strategies. This study aimed to determine the drug-susceptibility profiles of MAC clinical isolates and to investigate the genetic basis conferring drug resistance using whole-genome sequencing (WGS) analysis. Drug-susceptibility profiles based on minimum inhibitory concentration (MIC) assays were determined for 38 MAC clinical isolates (12 Mycobacterium avium and 26 Mycobacterium intracellulare). Mutations associated with drug resistance were identified through genome analysis of these isolates, and their phylogenetic relationships were also examined. Drug resistance, based on MIC values, was most commonly observed for moxifloxacin (81.6%), followed by linezolid (78.9%), clarithromycin (44.7%) and amikacin (36.8%). We identified specific mutations associated with resistance to amikacin. These include the rrs mutation at C464T in amikacin intermediate-resistance M. avium, and two mutations at T250A and G1453T in amikacin non-susceptible M. intracellulare. Mutations in rrl at A2058G, A2059C and A2059G were potentially linked to clarithromycin resistance. MAC clinical isolates not susceptible to linezolid exhibited mutations in rplC at G237C and C459T, as well as two rplD mutations at G443A and A489G. GyrB substitution Thr521Ala (T521A) was identified in moxifloxacin non-susceptible isolates, which may contribute to this resistance. A phylogeny of our MAC isolates revealed high levels of genetic diversity. Our findings suggest that the standard treatment regimen for MAC infections using moxifloxacin, linezolid, clarithromycin and amikacin may be driving development of resistance, potentially due to specific mutations. The combination of phenotypic and genotypic susceptibility testing can be valuable in guiding the clinical use of drugs for the treatment of MAC infections.

Toward a global virus genomic surveillance network

The COVID-19 pandemic galvanized the field of virus genomic surveillance, demonstrating its utility for public health. Now, we must harness the momentum that led to increased infrastructure, training, and political will to build a sustainable global genomic surveillance network for other epidemic and endemic viruses. We suggest a generalizable modular sequencing framework wherein users can easily switch between virus targets to maximize cost-effectiveness and maintain readiness for new threats. We also highlight challenges associated with genomic surveillance and when global inequalities persist. We propose solutions to mitigate some of these issues, including training and multilateral partnerships. Exploring alternatives to clinical sequencing can also reduce the cost of surveillance programs. Finally, we discuss how establishing genomic surveillance would aid control programs and potentially provide a warning system for outbreaks, using a global respiratory virus (RSV), an arbovirus (dengue virus), and a regional zoonotic virus (Lassa virus) as examples.

Global SARS-CoV-2 genomic surveillance: What we have learned (so far)

The COVID-19 pandemic has brought significant challenges for genomic surveillance strategies in public health systems worldwide. During the past thirty-four months, many countries faced several epidemic waves of SARS-CoV-2 infections, driven mainly by the emergence and spread of novel variants. In that line, genomic surveillance has been a crucial toolkit to study the real-time SARS-CoV-2 evolution, for the assessment and optimization of novel diagnostic assays, and to improve the efficacy of existing vaccines. During the pandemic, the identification of emerging lineages carrying lineage-specific mutations (particularly those in the Receptor Binding domain) showed how these mutations might significantly impact viral transmissibility, protection from reinfection and vaccination. So far, an unprecedented number of SARS-CoV-2 viral genomes has been released in public databases (i.e., GISAID, and NCBI), achieving 14 million genome sequences available as of early-November 2022. In the present review, we summarise the global landscape of SARS-CoV-2 during the first thirty-four months of viral circulation and evolution. It demonstrates the urgency and importance of sustained investment in genomic surveillance strategies to timely identify the emergence of any potential viral pathogen or associated variants, which in turn is key to epidemic and pandemic preparedness.

Incorporating variant frequencies data into short-term forecasting for COVID-19 cases and deaths in the USA: a deep learning approach

BACKGROUND

Since the US reported its first COVID-19 case on January 21, 2020, the science community has been applying various techniques to forecast incident cases and deaths. To date, providing an accurate and robust forecast at a high spatial resolution has proved challenging, even in the short term.

METHOD

Here we present a novel multi-stage deep learning model to forecast the number of COVID-19 cases and deaths for each US state at a weekly level for a forecast horizon of 1-4 weeks. The model is heavily data driven, and relies on epidemiological, mobility, survey, climate, demographic, and SARS-CoV-2 variant frequencies data. We implement a rigorous and robust evaluation of our model-specifically we report on weekly performance over a one-year period based on multiple error metrics, and explicitly assess how our model performance varies over space, chronological time, and different outbreak phases.

FINDINGS

The proposed model is shown to consistently outperform the CDC ensemble model for all evaluation metrics in multiple spatiotemporal settings, especially for the longer-term (3 and 4 weeks ahead) forecast horizon. Our case study also highlights the potential value of variant frequencies data for use in short-term forecasting to identify forthcoming surges driven by new variants.

INTERPRETATION

Based on our findings, the proposed forecasting framework improves upon the available state-of-the-art forecasting tools currently used to support public health decision making with respect to COVID-19 risk.

FUNDING

This work was funded the NSF Rapid Response Research (RAPID) grant Award ID 2108526 and the CDC Contract #75D30120C09570.